stealthmodels.

MODEL FIELD GUIDE

GLM-5.3-Flash (Ox Alpha): Ground Truth

Formerly Ox Alpha, this MIT-licensed multimodal MoE combines a million-token context with low hosted token rates. Read the Ox Alpha reveal and model guide →

Drawn by this model01 / 01
Ground Truth benchmark result generated by this model
SVG BENCHMARKGround TruthView result

A good fit for

Cost-conscious coding and multimodal agents through an API, or self-hosting where several hundred gigabytes of model memory are available.

Context window
1,000,000tokens[3]
Active parameters
18Bparameters[1]
License
MIT[2]
Model factsArchitecture, context & capabilities
First available
2026-08-26[3]
Developer
Z.AI[1] [2]
Family
GLM-5[1] [2]
Context window
1,000,000 tokens[3]
Maximum output
131,000 tokens[3]
License
MIT[2]
Architecture
Hybrid sparse and linear attention MoE[1]
Total parameters
320B[1]
Active parameters per token
18B[1]
Input
text, image[1]
Output
text[1]
API access & pricingOpenRouter · $0.15 input / $0.5 output · per 1M tokens

Vercel AI Gateway / DigitalOcean route; input 0.15 USD/1M; cached read 0.03/1M; output 0.5/1M; A specific gateway provider route; Vercel lists other providers with different prices and limits.

[3]
  • OpenRouterObserved 2026-10-06
    Input $0.15Output $0.5Cached read $0.03

    per 1M tokens

    z-ai/glm-5.3-flash[5]
  • Vercel AI Gateway / DigitalOcean routeObserved 2026-10-06
    Input $0.15Output $0.5Cached read $0.03

    per 1M tokens

    Available

    zai/glm-5.3-flash[3]
Run it locallyWeights, memory & deployment

multi-GPU workstation/server

[1] [4]
  • Unsloth UD-Q4_K_XL GGUF200 GB storage · ~4-bit[4]
Measurements & sources6 primary references · Speed & serving conditions
  • Vercel AI Gateway / DigitalOceanObserved 2026-10-06

    119 tokens/s decode

    500 ms to first token

    [3]
  • Artificial AnalysisObserved 2026-10-06

    53 tokens/s decode

    current

    [6]

StealthMark score

StealthMark measures fluid and visual intelligence of AI models.

General and Visual

General Intelligence

Perspective, anatomy, correct shadows, purposeful detail and physical consistency.

Visual Intelligence

Likeness, anatomy, detail, light, materials, composition and artistic expression.

Judging

Frontier Models blindly judge the AI generation unbiased through an intricate process. Our researchers control for mistakes.

Overall score

Overall combines weighted task scores on a scale from 0 to 100. Repeating a task does not increase its weight. Estimated means the total includes an estimate for one missing scene.

Benchmaxxing

Benchmaxx flags models that score unusually well on the familiar Pelican task compared with their other benchmark results. The percentage measures that imbalance, not the probability of training contamination. Flagged Pelican scores are excluded from Overall. Pelican never counts toward General or Visual.

We keep some prompts private to discourage test-specific optimization.

Scoring categories

Prompt & likeness

Subjects, actions and likeness.

Anatomy & construction

Coherent bodies, joints and machinery.

Space & placement

Perspective, scale and contact.

Light & reflections

Lighting, shadows and reflections.

Purposeful detail

Clear, purposeful small features.

Artistry & character

Composition, expression and drawing skill.

Materials & effects

Surfaces, texture and movement.

Visual integrity

Clean shapes, layers and edges.

BENCHMARK RESULTS

Compare results

This link opens the same results in the same order.

StealthMark benchmark results

Reset benchmark preferences?

Return to the default view, clear comparison selections and restore the advertisement.